Label-Free Oligonucleotide-Based SPR Biosensor for the Detection of the Gene Mutation Causing Prothrombin-Related Thrombophilia

Prothrombin-related thrombophilia is a genetic disorder produced by a substitution of a single DNA base pair, replacing guanine with adenine, and is detected mainly by polymerase chain reaction (PCR). A suitable alternative that could detect the single point mutation without requiring sample amplification is the surface plasmon resonance (SPR) technique. SPR biosensors are of great interest: they offer a platform to monitor biomolecular interactions, are highly selective, and enable rapid analysis in real time. Oligonucleotide-based SPR biosensors can be used to differentiate complementary sequences from partially complementary or noncomplementary strands. In this work, a glass chip covered with an ultrathin (50 nm) gold film was modified with oligonucleotide strands complementary to the mutated or normal (nonmutated) DNA responsible for prothrombin-related thrombophilia, forming two detection platforms called mutated thrombophilia (MT) biosensor and normal thrombophilia (NT) biosensor. The results show that the hybridization response is obtained in 30 min, label free and with high reproducibility. The sensitivity obtained in both systems was approximately 4 ΔμRIU/nM. The dissociation constant and limits of detection calculated were 12.2 nM and 20 pM (3 fmol), respectively, for the MT biosensor, and 8.5 nM and 30 pM (4.5 fmol) for the NT biosensor. The two biosensors selectively recognize their complementary strand (mutated or normal) in buffer solution. In addition, each platform can be reused up to 24 times when the surface is regenerated with HCl. This work contributes to the design of the first SPR biosensor for the detection of prothrombin-related thrombophilia based on oligonucleotides with single point mutations, label-free and without the need to apply an amplification method.


S1 Construction of biosensors with mutated or normal thrombophilia strands
To improve the response of each biosensor, all stages of modification of the gold surface were optimized.
Self-assembled monolayer formation: 150 L of 4MBA solution in ethanol was evaluated at 10 and 1.0 mM. The concentration of 4 MBA (1.0 mM) was chosen due to the greater reproducibility in the responses obtained for the activation stage using EDC/NHS. Activation of carboxylic groups: One or two injections of EDC/NHS (0.2 M and 0.05 M, respectively) were evaluated. Two injections of EDC/NHS were used since the first and second injections both registered increases in the activation response of 545 ± 14 RU and 212 ± 25 RU, respectively (n =10). Consequently, using two injections, a greater response was recorded in the immobilization of aminated strands.
Strand immobilization: Two types of buffer solutions were evaluated, PBS and Tris-EDTA, at pH 7.4, with flow rates of 5, 10 and 20 L/min, ionic strength of 0.1 or 0.3 M, and Tween20 0.25% v/v. At a 1.4 M concentration of MT-A or NT-A strands, Tris-EDTA buffer was used at 5 L/min for 50 min, without ionic strength and without Tween20. The selection criterion was the highest response obtained due to the greatest number of immobilized strands, with 137 ± 10 RU and 122 ± 11 RU for the immobilization of MT-A and NT-A, respectively (n =10).
Blocking: An injection of 1.0 M ethanolamine was compared with successive injections (between 1 and 5) of 0.1 M ethanolamine at pH 8.5, adjusted with HCl. Three consecutive injections of ethanolamine 0.1 M at pH 8.5 were used due to the higher response obtained in the working channel and lower response in the reference channel during the hybridization stage with complementary strands MT-C or NT-C.

S2 Binding site calculations
All data were obtained considering the molecular weights of aminated and complementary strands described in Table 2, Section 2.2 of the Materials and Methods. Table S1 shows the immobilization response of MT-A or NT-A and the surface density and binding sites calculated using the respective equations.

Surface Density (g/cm 2 ) Binding Site (mol/cm 2 )
MT Biosensor 137 ± 10 1.37×10 -8 ± 1.0×10 -9 2.56×10 -4 ± 1.87×10 -5 NT Biosensor 122 ± 11 1.22×10 -8 ± 1.1×10 -9 2.26×10 -4 ± 2.03×10 -5 S3 Reuse of each biosensor Figure S1 and S2 show 24 consecutive injections of complementary strands at a concentration of 40 nM into the corresponding biosensor, maintaining over 95% of the initial response.   Table S2 shows the concentrations, average responses and standard deviations obtained for injections of complementary strands MT-C (in the MT biosensor) and NT-C (in the NT biosensor).  Table S3 shows the equations used and parameters obtained for the response versus concentration of complementary strands in the MT biosensor. The first adjustment was a Langmuir isotherm model for the responses obtained from MT-C between 0.5 and 100 nM. Then, a linear fit was used in the range between 0.1 and 1.0 nM.  Table S4 shows the equations used and parameters obtained for the response versus concentration of complementary strands in the NT biosensor. The first adjustment was a Langmuir isotherm model for the responses obtained from NT-C between 0.1 and 10 nM. Then, a linear fit was used in the range between 0.1 and 0.8 nM.  Table S5 shows the values obtained for KD and Ka and the theoretical response maximum and ligand activity for both biosensors using the following equations: